The Mitochondrial Protein VDAC1 at the Crossroads of Cancer Cell Metabolism: The Epigenetic Link.

Carcinogenesis is a complicated process that involves the deregulation of epigenetics, resulting in cellular transformational events, such as proliferation, differentiation, and metastasis. Most chromatin-modifying enzymes utilize metabolites as co-factors or substrates and thus are directly dependent on such metabolites as acetyl-coenzyme A, S-adenosylmethionine, and NAD+. Here, we show that using specific siRNA to deplete a tumor of VDAC1 not only led to reprograming of the cancer cell metabolism but also altered several epigenetic-related enzymes and factors. VDAC1, in the outer mitochondrial membrane, controls metabolic cross-talk between the mitochondria and the rest of the cell, thus regulating the metabolic and energetic functions of mitochondria, and has been implicated in apoptotic-relevant events. We previously demonstrated that silencing VDAC1 expression in glioblastoma (GBM) U-87MG cell-derived tumors, resulted in reprogramed metabolism leading to inhibited tumor growth, angiogenesis, epithelial-mesenchymal transition and invasiveness, and elimination of cancer stem cells, while promoting the differentiation of residual tumor cells into neuronal-like cells. These VDAC1 depletion-mediated effects involved alterations in transcription factors regulating signaling pathways associated with cancer hallmarks. As the epigenome is sensitive to cellular metabolism, this study was designed to assess whether depleting VDAC1 affects the metabolism-epigenetics axis. Using DNA microarrays, q-PCR, and specific antibodies, we analyzed the effects of si-VDAC1 treatment of U-87MG-derived tumors on histone modifications and epigenetic-related enzyme expression levels, as well as the methylation and acetylation state, to uncover any alterations in epigenetic properties. Our results demonstrate that metabolic rewiring of GBM via VDAC1 depletion affects epigenetic modifications, and strongly support the presence of an interplay between metabolism and epigenetics.


Immunohistochemistry (IHC)
Immunohistochemical staining was performed on formalin-fixed and paraffin-embedded tumors obtained from si-NT-and si-hVDAC1-treated tumors as described previously [1]. Sections were deparaffinized using xylene and a graded ethanol series. Endogenous peroxidase activity was blocked by incubating the sections in 3% H2O2 for 10 minutes. Antigen retrieval was performed in 0.01M citrate buffer (pH 6.0) at 95-98 °C for 20 min. After washing sections in PBS (pH 7.4), nonspecific antibody binding was reduced by incubating the sections in 10% normal goat serum for 2 h. After decanting excess serum, sections were incubated overnight at 4 °C with primary antibodies (Table S1). After washing with PBS, the sections were incubated for 2 h with the appropriate secondary antibodies conjugated to horseradish peroxidase (Table S1). Sections were washed three times in PBS and subsequently, the peroxidase-catalyzed reaction was visualized by incubating with 0.02% DAB. After rinsing in water, the sections were counterstained with hematoxylin, and mounted with Vectashield mounting medium (Vector Laboratories, Burlingame, CA, USA). Finally, the sections were observed under a microscope (DM2500, Leica, Wetzlar, Germany) and images were captured at the indicated magnification with the same light intensity and exposure time. Controls were carried out with the same protocols but omitting the primary antibodies.

RNA Preparation and DNA Microarray Analysis
Total RNA was isolated from si-Scr-and si-hVDAC1-treated tumors (4-6 mice each) using the RNeasy mini kit (Qiagen) according to the manufacturer's instructions. Total RNA quality was analyzed using the Agilent RNA 6000 nano kit. The RNA integrity values obtained for total RNA extracted from si-Scr-and si-hVDAC1-treated tumors were 8 to 10. The targets for Affymetrix whole transcript expression microarray analyses were prepared using the Affymetrix GeneChip WT PLUS reagent kit according to the manufacturer's instructions and hybridized to Human Gene 1.0 ST microarrays. Data were acquired using the Affymetrix GeneChip algorithm (version 3.2). Files were imported to Partek Genomics Suite, and all probes except control probes were pre-processed by RMA background correction, log2 transformation and probe set summarization using median polish. Probe sets with signals < 5 in all samples were filtered out. Subsequently, global scaling was carried out by shifting the mean of each sample to the grand mean (i.e. in each sample, the mean signal was subtracted from each of the signals, and then the grand mean of all the samples was added, such that all arrays eventually had the same mean signal). Differentially expressed genes were defined as those having FDR-adjusted t test p-value < 0.05, and two clusters were defined, up-regulated and downregulated genes (linear fold change > 2 or < -2, respectively). Each cluster was tested separately for enrichment of functional groups based on the GO system.

Quantitative Real-Time PCR (q-RT-PCR)
Real-time RT-PCR was performed (KiCqStart Primers; Sigma Aldrich, St. Louis, Missouri, USA) in triplicate, using the Power SYBR green master mix (Applied Biosystems, Foster City, CA, USA). Genes examined and primers used are listed in Table S2.Levels of target genes were normalized relative to -actin mRNA levels. Samples were amplified by a 7300 Real Time PCR System (Applied Biosystems) for 40 cycles using the following PCR parameters: 95 °C for 15 seconds, 60 °C for 1 minute, and 72 °C for 1 minute. The copy numbers for each sample were calculated by the CT-based calibrated standard curve method. The mean fold changes (± SEM) of the three replicates were calculated.

Gel Electrophoresis and Immunoblotting
To extract proteins for immunoblotting, tumor tissues were solubilized in a lysis buffer (50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 1 mM EDTA, 1.5 mM MgCl2, 10% glycerol, 1% Triton X-100, a protease inhibitor cocktail (Calbiochem)), followed by sonication and centrifugation (10 min, 600 g). The protein concentration of each lysate was determined using a Lowry assay. Samples were stored in −80 °C until analysis by gel electrophoresis and immunoblotting, as described in Supplementary Data.
For immunostaining, membranes containing electro-transferred proteins following SDS-PAGE were blocked with 5% non-fat dry milk and 0.1% Tween-20 in TBS, incubated with the primary antibodies (sources and dilutions as detailed in Table S1) and then with HRP-conjugated anti-mouse or anti-rabbit (1:10,000) or anti-goat (1:20,000) IgG. Enhanced chemiluminescent substrate (Pierce Chemical, Rockford, IL, USA) was used to detect HRP activity. Band intensities were analyzed by densitometry using FUSION-FX (Vilber Lourmat, France) software, and values were normalized to the intensities of the appropriate -actin signal that served as a loading control.

LC-HR MS/MS Analysis
Tumors from three mice treated with si-NT or si-hVDAC1 were subjected to in-solution tryptic digestion as follows. Proteins were first reduced by incubation with 5 mM DTT for 30 min at 60 °C, followed by alkylation with 10 mM iodoacetamide in the dark for 30 min at 21 °C. Proteins were then subjected to digestion with trypsin (Promega, Madison, WI, USA) at a 1:50 trypsin:protein ratio for 16 h at 37 °C. Following digestion, detergents were cleared from the samples using commercial detergent removal columns (Pierce, Rockford, IL, USA), and desalted using solid-phase extraction columns (Oasis HLB, Waters, Milford, MA, USA). Digestions were stopped by addition of trifluroacetic acid (1%). The samples were stored at −80 °C until LC-HR MS/MS analysis.
For LC-HR MS/MS, ULC/MS grade solvents were used for all chromatographic steps. Each sample was separated using split-less nano-ultra performance liquid chromatography columns (10K psi nanoAcquity; Waters). The mobile phase was (A) H2O and 0.1% formic acid, and (B) acetonitrile and 0.1% formic acid. Desalting of the samples was performed online using a reverse-phase C18 trapping column (180 µm internal diameter, 20 mm length, 5 µm particle size; Waters). The peptides were then separated on a T3 HSS nano-column (75 µm internal diameter, 250 mm length, 1.8 µm particle size; Waters) at 0.3 µL/min. Peptides were eluted from the column into the mass spectrometer using the following gradient: 4% to 35% (B) for 150 min, 35% to 90% (B) for 5 min, maintained at 90% for 5 min and then back to initial conditions. The nano-UPLC was coupled online through a nano-ESI emitter (10 µm tip; New Objective, Woburn, MA, USA) to a quadrupole Orbitrap mass spectrometer (Q Executive, Thermo Scientific, Waltham, MA, USA) using a FlexIon nanospray apparatus (Proxeon, Odense, Denmark). Data was acquired in the DDA mode, using a Top12 method [2]. Raw data was imported into Expressionist software (Genedata, Switzerland) [3,4]. The software was used for retention time alignment and peak detection of precursor peptide intensities. A master peak list was generated from all MS/MS events and sent for database searching using Mascot v2.4 (Matrix Sciences, Maharashtra, India). Data was searched against a database containing forward and reverse human protein sequences from UniprotKB/SwissProt, and 125 common laboratory contaminants, totaling 20,304 entries. Fixed modification was set to carbamidomethylation of cysteines, while variable modification was set to oxidation of methionines. Search results were then imported back to Expressionist for annotation of detected peaks. Identifications were filtered such that the global false discovery rate was a maximum of 1%. Protein abundance was calculated based on the three most abundant peptides [5]. Proteins with less than 2 unique peptides were excluded from further analysis.
LC-HR-MS/MS data were imported into Partek Genomics Suite software (Partek, St. Louis, MO, USA) and differences between expression levels of the proteins in the different groups were calculated using a t-test. Functional enrichment analysis of differentially expressed proteins was performed using the DAVID and Gene Ontology (GO) bioinformatics resources, v6.7 [6]. Figure S1. Volcano plot of genes with altered expression. Volcano plot of all genes showing p-value as function of the fold change in the si-hVDAC1-TTs relative to si-NT-TTs. Cutoff lines are at FDRadjusted p-value = 0.05 and linear fold change of decreased or increased expression ≥ −2 and 2, respectively in the si-hVDAC1-TTs relative to si-NT-TTs. Total number of differentially expressed genes was 5,271 of which 2,291 were down-regulated transcripts and 2,980 were up-regulated transcripts.    Table S2. Primers for Q-RT-PCR used in this study. The genes examined, and the forward and reverse sequences of the primers used are indicated.